Development of normal retinal organization depends on Sonic hedgehog signaling from ganglion cells

Article metrics


The adult retina is organized into three cellular layers—an outer photoreceptor, a middle interneuron and an inner retinal ganglion cell (RGC) layer. Although the retinal pigment epithelium (RPE) and Müller cells are important in the establishment and maintenance of this organization1,2, the signals involved are unknown. Here we show that Sonic hedgehog signaling from RGCs is required for the normal laminar organization in the vertebrate retina.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Figure 1: Shh signaling restores lamination and Müller cell morphology in retinal explants.
Figure 2: Inactivation of Shh expression in the retina reduces Hh target gene expression and results in lamination defects.
Figure 3: Immature photoceptor cells participate in rosette formation in Thy1-Cre;Shh−/c mice.


  1. 1

    Raymond, S.M. & Jackson, I.J. Curr. Biol. 5, 1286–1295 (1995).

  2. 2

    Willbold, E. & Layer, P.G. Histol. Histopathol. 13, 531–552 (1998).

  3. 3

    Jensen, A.M. & Wallace, V.A. Development 124, 363–371 (1997).

  4. 4

    Goodrich, L.V. & Scott, M.P. Neuron 21, 1243–1257 (1998).

  5. 5

    Lillien, L. & Cepko, C. Development 115, 253–266 (1992).

  6. 6

    Chiang, C. et al. Nature 383, 407–413 (1996).

  7. 7

    Campsall, K.D., Mazaerolle, C.J., de Repentingy, Y., Kothary, R. & Wallace, V.A. Dev. Dynamics 224, 135–143 (2002).

  8. 8

    Lewis, P.M. et al. Cell 105, 599–612 (2001).

  9. 9

    Wallace, V.A. & Raff, M.C. Development 126, 2901–2909 (1999).

  10. 10

    Furukawa, T., Morrow, E.M. & Cepko, C.L. Cell 91, 531–541 (1997).

  11. 11

    Zhang, X.M., Ramalho-Santos, M. & McMahon, A.P. Cell 106, 781–792 (2001).

  12. 12

    Neumann, C.J. & Nuesslein-Volhard, C. Science 289, 2137–2139 (2000).

  13. 13

    Brown, N.L., Patel, S., Brzezinski, J. & Glaser, T. Development 128, 2497–2508 (2001).

  14. 14

    Dahmane, N. & Ruiz-i-Altaba, A. Development 126, 3089–3100 (1999).

  15. 15

    Hatten, M.E. Annu. Rev. Neurosci. 22, 511–539 (1999).

Download references


We thank R. Kothary, D. Picketts and M. Raff for reading the manuscript. Recombinant Shh protein was a gift from D. Baker, Biogen. V.A.W. is supported by the National Cancer Institute of Canada with funds from the Terry Fox Run, the Canadian Institutes of Health Research (CIHR), the Multiple Sclerosis Society of Canada and the EA Baker Foundation for the Prevention of Blindness. A.P.M. is supported by National Institutes of Health grant NS 33642. V.A.W. is the recipient of a CIHR New Investigator Award.

Author information

Correspondence to Valerie A. Wallace.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Supplementary information

Supplementary Fig. 1.

Maintenance of Hh target gene expression requires RGC-derived Shh signaling. (a) In situ hybridization for Shh and Gli1 mRNA and immunohistochemistry for Brn3b1, a RGC marker, or TUNEL staining on explant sections of mouse E18 neural retina. At time 0, Shh and Brn3b are expressed in RGCs and Gli1 is expressed in the neuroblast layer. By 24 hours, RGCs in control and Shh-N-treated explants are dying, as revealed by TUNEL staining and the loss of Shh expression and Brn3b immunoreactivity. The loss of RGCs and Shh expression also correlates with downregulation of Gli1 expression. Progenitor cells in retinal explants maintain their competence to respond to Hh signaling, as treatment with recombinant Shh-N restores Gli1 expression in the neuroblast layer. (b) Semi-quantitative RT-PCR analysis of gene expression in the retina at E18 and in retinal explants cultured for 24 or 48 hours in the presence or absence of recombinant Shh-N. The level of Shh mRNA is reduced after 24 hours of culture, which correlates with a reduction in the levels of Gli1 and Ptc 2 (Ref 2) mRNA in untreated explants. Shh-N-treatment restores the levels of Gli1 and Ptc2 mRNA without affecting the level of Shh mRNA. The levels of Ptc1 mRNA do not change appreciably under any of the treatment conditions. Ihh is not expressed in retinal explants, although its expression is detected in the retinal pigment epithelium (RPE). Smo is expressed in retinal explants under all treatment conditions. Amplification of GAPDH was used as a control for equal loading and reactions carried out without reverse transcriptase were used as a control of genomic DNA contamination (not shown). (JPG 62 kb)

Supplementary Fig. 2.

Shh treatment restores normal lamination and Müller cell morphology in retinal explant cultures. (a, b) Explants of mouse E18 neural retina were cultured for 7 days in the absence of growth factors (control), or in the presence of Shh-N (2μg/ml), EGF (10ng/ml), anti-Hh or anti-LFA3 antibodies (each at 30 μmg/ml). Hoescht and anti-rhodopsin staining reveal the presence of photoreceptor rosettes in the outer layer of control, EGF, anti-Hh and anti-LFA-3-treated explants, whereas rosetting is markedly reduced in Shh-N treated explants. Müller cells, identified with anti-CRALPB antibodies, are disorganized in control, EGF, anti-Hh and anti-LFA3-treated explants, whereas they adopt a radial morphology in Shh-N-treated explants. (b) Examples of rhodopsin and CRALBP staining in control, Shh-N and anti-Hh-treated explants photographed at higher magnification. The majority of cells in the rosettes (control and anti-Hh-treated explants) and the outer nuclear layer (Shh-N-treated explant) are rod photoreceptor cells. Müller cells in Shh-N-treated explants extend apical processes into the photoreceptor layer, whereas this rarely occurs in control or anti-Hh-treated explants. Nuclei were counterstained with Hoescht. Gliosis, identified with anti-GFAP antibodies is reduced in Shh-N-treated compared with control explants after 14 days in culture. The anti-Hh monoclonal antibody, 5E1, reacts with both Shh and Indian hedgehog3, however, Ihh is not expressed in retinal explants (Supplementary Fig. 1), thus the effect of the anti-Hh treatment in retinal explants is due to inhibition of Shh activity. (c) The extent of normal lamination in explants grown under the indicated conditions for 7 days was quantified by determining the length of the non-rosetted regions as a function of the total length of each explant section. Experiment 1, n = 4 and Experiment 2, n = 2 (control) and n = 3 explants/condition. Measurements were performed on at least 3 sections/condition. (JPG 104 kb)

Supplementary Fig. 3.

Retinal progenitor cells participate in rosette formation in Thy1-Cre;Shh-/c mice. (a-f) Retinal sections from wild type (a, b) and Thy1-Cre;Shh-/c (c-f) littermates at E17 processed for in situ hybridization for the expression of the retinal progenitor cell markers CyclinD1 and Chx10. (e, f) Higher magnification view of the boxed areas in (c, d), respectively. (JPG 43 kb)

Supplementary Methods (PDF 47 kb)

Rights and permissions

Reprints and Permissions

About this article

Further reading